Did some nodes drop?  I've lost connection along with 7 or 8Phs of others.

Edit: looks like the main node and the NYA nodes are down.  I'm back up on the Oregon one now.

Edit 2: The main and New York nodes just came back.

And I now have all of the nodes in my failover list.

Both are running 725Mhz, one power level 5 and the other power level 6 for a total of 1.94THs.

My two R606's are pulling 207 Watts, including the Power Supply and minus the Pi 4.

Well, I may not have known what I was talking about with the blockerupter compile switch, which I said at the time.

But I know for a fact that my two R606 Miners run MUCH better on Ubuntu Linux and on the Pi 4 with their raspian distro.  I messed around with Win7 on two different machines, one of which also has the Ubuntu install, I have verified that I could not, in no way, have gotten 1Ths out of either R606 on Win7. But on the Pi 4 and my Ubuntu machine its no problem at all.

I had mined a long time with the ASIC Miner Block Erupters, and my BFL Jalapeno, and would have constant problems with the 78 Block Erupters.  Now I'm wondering if it was the Win7 USB bottleneck?  I'm thinking it was.

So, if your thinking your going to get the same USB performance out of Windoze as with Linux, Your Wrong!  And I mean that as kindly as can be.  Just go back to May of last year when we first got our R606's and witness our issues with Win7 and Win10.  Poor VH finally discontinued support for Windoze after working his but off trying helping us.  -ck, aka "the developer", had dropped support for Windoze quite a while before that.

You need to read the CGminer readme files and check the CGMiner thread.

I know very little about compiling the source.  But there is no "--enable-blockerupter" option as it uses the icarus driver, there is no need to disable gpu support as that was removed in 4.8, I beleive, and I have never seen an "disable" switch so I don't think there is any need for that.

Yes, I use a Pi4 for my two R606's.  No hub needed for me.

I wasted allot of time dinking around with a couple Win7 machines.  It was slightly fun and educational.  But I love my Pi4.

Mine says 856.  Sidehack has to work on his fortune cookie messages

Just a WAG, but it sounds like you heat damaged it with the filter restricting air flow?  You said it was really hot.

My batch 2 unit was happy at 750 for a long time, then chip 4 would drop out every morning and then come back up.  So I dropped my target to 725 and it seems fine now and mine never got hot.  But I may have to change my expectations again down the road.  The stock is 550Mhz after all

I feel like I have gotten good mileage out of my two units.

You need to look through the original CGMiner thread.  CK made some arguments for this purpose.  It's been too many years for me to remember them off hand. But it was something like --GSI:0.  Where GSI is the R606 and the :0 means to use none of them.  So once you find the correct command line argument that would be what you add to the Newpac instance and then do the same thing to not use the Newpacs in the R606 session.

Well you may be driving them too hard.  I tested one of my R606's at 800 to 850 Mhz and found that, although it would run stable at that frequency, that it would consume more power and actually have a lower hash rate than 750Mzh.  At 750 target it started dropping chip 4 once in a while and reset to get it back.  So I have dropped it to 725Mhz and it doesn't do that anymore.

So I would drop the target frequency down.  I would never go above 750Mhz even if a unit ran higher freq's.  So drop the target down to a frequency where you don't loose those chips, probably 700 to 750 range.

And don't just randomly raise your voltage to the max as that can be counter productive too.  Take your voltage up one step at a time and do test runs at each step.

Changing diff to higher than required to find a block doesn't sound like a good idea.

Any decent pool has the ability to adjust miner difficulty to around 18 shares per minute.  So you don't need to mess with it yourself.

You didn't say what instructions you used to install CGMiner.  So here is a link to detailed instructions to review to see if you missed anything.

https://bitcointalk.org/index.php?topic=5053833.msg50968875#msg50968875

Also try with just one stick.  After you get one working add the others one at a time.  You may have a power issue on your hub.

That's my .02BTC worth of suggestions.

Edit:


I just noticed that you have 2pac's.  They have a separate thread from the R606's.  Not sure how different they are but you should post in that thread.

No blocks were found for those months.

If we found a block right now it would read 2.42 blocks expected for April since we are at 242% of current difficulty now.

Would that be functionally different from Ubuntu 18.04?

Edit: Isn't Ubuntu a Debian distro?

Maybe bump your power level up to 5 after getting the 85f version of CGMiner.

Also keep in mind the factory burn in was done on a Ubuntu Debian Linux machine.  That does make a big difference.  I think you can install Ubuntu along side of a Win10 installation.

I tested with my laptop which I setup to dual boot to Win7-Ubuntu-OS/2.  So I was a able to do objective testing on the same hardware with Win7 and Ubuntu 18.04.  Under Ubuntu I was able to go up to 800Mhz, 1.012Ths, which I would not recommend, but it was stable.  Win7 only went up to 668.75Mhz for 838GHs.

I have two R606's, whats the difference between and R606 and a P606?  How did you fix it with the NewPacs?


Mine did that on Win7 when I was going through a USB 2.1 Hub.  I quit using the Hub, I quit using Windoze and ended up with a Pi4 and that fixed everything for me.  Also I never used a brick Power Supply, I used a Computer Power Supply and plugged into the PCIE Port.

Edit: If your hell bent on using Winblows try the "May 13, 2019 @ a62385f" version of CGMiner not the latest one.  That one would do about 631.25Mhz at 804GHs on Power Level 5 for me on my Win7 machine.

The Pi4 runs the two of mine at 725 Mhz for 1.95Ths total.  Again you wont get that rate with the Brick Power Supply.

So the share difficulty will eventually adjust down?

If that is so will the pool still honor the "--suggest-diff" argument?

I, for one, like this change.

Run it in a DOS window so that you can see what the error message is.

That is correct.

All of your hashrate is combined on a single mission, regardless of how many addresses you distribute it too or how many devices you consolidate to one address..  Only your total hash rate matters.  The more the better chance of finding a block.